Introduction
Atypical hemolytic uremic syndrome (aHUS) is a very rare, life-threatening thrombotic microangiopathy mainly caused by uncontrolled complement activation with a reported incidence of 0.5 per million per year [
1]. Approximately, 50–70% of the patients have causative mutations in genes encoding complement factor H (20–30%), membrane cofactor protein (
MCP; 10–15%), complement factor I (4–10%), factor B (1–2%), C3 (5–10%), thrombomodulin (5%), or diacylglycerol kinase ɛ (3%), while 6% of aHUS patients have anti-factor H antibodies [
2‐
4]. The underlying genetic or immunologic cause is thought to determine the clinical outcome. While patients with mutations in complement factor H suffer mostly from a severe clinical course, patients with an isolated
MCP mutation have a more favorable outcome despite reported relapses in up to 90% [
5‐
7]. Approximately, 80% of the pediatric patients with
MCP mutations show no long-term kidney impairment [
5,
7]. In an Italian cohort of MCP-aHUS children, none of the patients (0/13) progressed to chronic kidney disease (CKD) stage 5 or had CKD with proteinuria at 1- or 3-year follow-up [
7,
8]. In a French series with longer follow-up periods, 0% (0/12) of the children with MCP-aHUS progressed to CKD stage 5 at 1-year follow-up, while 17% (2/12) or 25% (3/12) showed CKD stage 5 at 5 years or median last follow-up of 17.8 years, respectively [
5].
Prior to the availability of effective treatment, aHUS patients had a very poor prognosis as up to 78% of the patients died or developed kidney failure within a few years [
3,
5,
7‐
10]. Treatment options were limited to plasma exchange or plasma infusion therapy. Since 2011, eculizumab, a monoclonal antibody targeting complement factor C5, has revolutionized the management of aHUS patients [
11,
12]. Eculizumab binds to the C5 protein thereby blocking its cleavage into the proinflammatory C5a and the pro-thrombotic C5b, which is a precursor of the lytic terminal complement complex. It is highly effective for treating acute episodes and preventing relapses, but needs to be administered intravenously at frequent intervals for maintenance therapy [
11]. In 2016, an international consensus approach for the management of aHUS in children recommended long-term eculizumab treatment for aHUS patients with a good response to treatment independently of the underlying mutation [
3].
Over the last years, discontinuation of long-term eculizumab treatment has been discussed controversially, but observational studies and guidelines are missing [
13‐
18]. The main rationale for discontinuation of eculizumab is to reduce the risk of severe side effects and high treatment costs. Side effects include the higher risk for meningococcal infections and complications due to intravenous injections and potential long-term immune-mediated drug reactions, even though this has so far not been observed [
19‐
21]. Furthermore, infusions twice a month have a high impact on the patients’ quality of life. Therefore, discontinuation of the recommended therapy is considered by many families and pediatric nephrologists, especially for children with
MCP mutations who have a reported low overall risk of kidney failure.
Here, we describe the initial aHUS presentation, timing of relapses, treatment, and kidney outcome of 20 children from German pediatric nephrology centers with aHUS due to isolated heterozygous MCP mutations. The objective of the present retrospective study was to report on the clinical presentation and to compare the outcome of patients receiving or not receiving eculizumab.
Discussion
Here, we describe one of the largest pediatric cohorts of patients suffering from aHUS due to isolated
MCP mutations. Most aHUS studies have included both adult and pediatric patients and have analyzed mutations in a variety of genes together [
32,
33]. However, it appears likely that different underlying genetic causes have variable disease outcomes. Furthermore, a previous study found that pediatric patients with aHUS due to
MCP mutations have a more favorable long-term kidney outcome compared with adult patients [
5].
In contrast to previous data [
5,
34,
35], males were predominant in our cohort. This could be explained by the fact that other groups have reported on patients with all genetic mutations, not focusing on patients with isolated
MCP mutation. However, larger case studies are needed to validate this finding. The mean age of onset in our cohort (4.6 years) was comparable with that reported in the cohort of the global (3.8 years) [
35] and Turkish (4.8 years) [
36] aHUS registries published in 2018, but higher compared with the mean age of onset published by the French registry in 2013 (1.5 years) [
5].
Despite the fact that kidney function appears to be well preserved in children with
MCP mutations, variable relapse rates of 20 [
7] up to 92% [
5] after a median follow-up time of 3 and 17.8 years, respectively, have been described. In our cohort, Kaplan-Meier analysis showed a relapse-risk rate of 7%, 29%, and 40% at 1, 3, and 5 years after first manifestation, respectively. It remains unclear why the French study showed much higher relapse rates [
5]. One explanation might be the longer follow-up period or the higher amount of homozygous
MCP mutations within the cohort of Fremeaux-Bacchi et al. (where 5 out of 12 patients had homozygous mutations) [
5]. The number of homozygous mutations within their cohort could also explain the earlier age of onset compared with our own cohort (1.5 vs. 4.58 years) [
5]. Additionally, different ethnic backgrounds could explain the variable clinical courses. The patients with relapses had a trend towards earlier onset of the disease compared with non-relapsing patients, but also a longer mean observation period. Therefore, additional studies with larger case numbers and longer follow-up periods are needed to further study the relapse rate of pediatric patients with
MCP mutations.
The overall kidney outcome in our cohort was good with a median eGFR above 100 ml/min/1.73m
2 at last follow-up. Only one patient developed CKD stage 5, 5.6 years after onset of aHUS. This is comparable with the published data demonstrating that
MCP mutations cause milder kidney damage than other forms of genetic aHUS, but still can lead to kidney failure [
35].
Currently, the recommended therapy for most aHUS patients is long-term administration of eculizumab, with a maintenance-dosing interval twice a month [
3]. The efficacy of this treatment has been shown in several prospective clinical trials, performed initially in adults and later in children [
11,
14,
37,
38]. However, discontinuation of eculizumab has been controversially discussed, as eculizumab has also potential adverse effects, especially with long-term treatment [
15‐
18,
21,
39]. In our cohort, only 4 out of 20 patients received long-term and 3 out of 20 patients temporary eculizumab treatment, which is explained partly by the onset of aHUS before the era of this treatment option.
As long-term eculizumab therapy is expensive and can lead to complications, we feel the role of eculizumab long-term treatment in patients who may never develop kidney failure or relapses needs to be considered carefully. Considering the low overall risk of kidney failure and a relapse risk of 10% per year, we personally are in favor of short-term eculizumab treatment.
Many centers already choose not to follow the recommended long-term regimen, especially for patients with underlying
MCP mutations and thus low risk of CKD stage 5. Wijnsma et al. have developed a treatment protocol of a restrictive eculizumab therapy for aHUS patients in the Netherlands [
18]. They recommend that irrespective of the underlying mutation, all aHUS patients should receive eculizumab, but after 3 months treatment should be re-evaluated and the drug should either be withdrawn or tapered [
18]. In our opinion, risks and benefits of long-term vs. short-term/on-demand eculizumab treatment have to be considered carefully in each individual case. While receiving eculizumab, the risk of meningococcal infection is increased more than 2000-fold compared with the healthy population [
21,
40,
41]. As eculizumab has only been used for aHUS since 2009, long-term and rare adverse effects have only systematically been analyzed over a period of 5 years from the global aHUS registry [
20]. After administration of eculizumab in the acute phase, it is difficult to motivate patients and parents in stable remission to undergo indefinite treatment without evidence of substantial benefit from prospective clinical trials. In our view, a short-term/on-demand eculizumab therapy, as proposed in the Netherlands, is probably sufficient for the majority of patients with
MCP mutations. The pre-requisite for this approach has to be a well-educated and compliant patient and family who are able to dipstick the urine regularly at home and measure blood pressure. In case of any sign of a potential relapse (microhematuria, proteinuria, oliguria, or potentially aHUS-triggering infections), the patient will have to be seen by a pediatrician immediately and eculizumab should be applied in case of relapse under pediatric nephrology care. As suggested by Wijnsma et al., we propose a re-evaluation of the treatment regimen at earliest after 3 months [
18]. In case of remission (normalized kidney function and hematological parameters), and depending on the amount of proteinuria, a withdrawal or tapering of eculizumab therapy should be taken into consideration. However, the current data on discontinuation of eculizumab is based on retrospective data from clinical case reports [
13] and analyses, as summarized above
. Also, we could not show a clear benefit from complement blockade therapy, independently of short- or long-term treatment. However, this may have been due to a small treatment group and high loss of follow-up in the non-relapsing group. Thus, independent prospective and larger observational studies or randomized controlled trials for pediatric and adult patients with isolated
MCP mutations are needed. Currently, a prospective observational study (NCT02574403) for all aHUS patients (> 3 years of age) to define the outcome after treatment discontinuation has been completed and results are expected shortly. This will help to predict kidney outcome and identify the group of patients who would benefit from long-term vs. short-term eculizumab treatment. Furthermore, potential relapse triggers and diagnostic tools for early detection of relapses should be identified.
Limitations of our study are the retrospective survey-based approach and the variable observation periods including a large number of patients who were lost to follow-up, especially in the non-relapsing group. A potential explanation for this observation was that patients without relapsing disease rather tended to miss follow-up visits compared with those with relapse experience. Even though a longer follow-up would have improved data quality, the Kaplan-Meier survival analysis censors drop-out patients and shows the relapse likelihood.
In summary, our retrospective analysis showed that without preventive therapy, children with isolated heterozygous MCP mutations have about a 40% risk of first relapse over 5 years. With the limitation of small treatment groups, there was no significant difference in long-term kidney function in the groups with vs. without and long- vs. short-term eculizumab treatment. Nevertheless, in our opinion, it appears safe to discontinue initial eculizumab treatment in individual patients in this subgroup, with an immediate restart of eculizumab in case of relapse. We think that larger observational studies are urgently needed, especially for patients with MCP mutations, in order to minimize the risks of the disease and its therapy while optimizing the treatment benefits.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.